Creating new objects

a <- 1 + 2
b <- 3 + 4
c <- "objects can be of different types"
c

[1] "objects can be of different types"

(d <- "another new string")

[1] "another new string"

a + b

[1] 10

How to name new objects

• Object names should be descriptive, i.e. allow you to quickly ascertain (or remember) what the object contains.
• They must start with a letter and can only contain letters, numbers, underscores and full stops
• You will need a way of naming objects using multiple words (no spaces allowed); here are a few options:
  • snake_case (used in most of the materials for this course)
  • camelCase (used in the gapminder examples that you will be working with as an exercise later)
  • you.can.also.use.full.stops
• Avoid spaces when naming variables (and when naming files) – it will make your life easier!

Code completion

• RStudio’s code completion function is useful for selecting an object that you want to inspect
• Create the following object:

this_is_a_really_long_name <- 2.5

• To try out code completion, type this, press Tab and see what happens
• If the start of the object name is not unique, you will need to add more letters / select from the list
• Create another object called this_is_too and then try code completion again

Going back to previous commands

• Let’s say you want to change the value of this_is_a_really_long_name to 3.5
• You can go back through the history of the commands typed into the console by pressing the “up” arrow key (Note: this only works with commands typed directly into the console)
• Select the command you want and then change it
• You can also search through the commands that you have run by typing this and pressing Cmd/Ctrl + up_arrow
• Use one of these methods to change the value of this_is_a_really_long_name

Be precise with object names

r_rocks <- 2^3
colour <- "blue"

• If you try to inspect these objects but aren’t completely precise in your spelling, you will get an error message
• For example, see what happens if you type in r_rock or R_rocks or if you use the American spelling of color
• Typos matter, case matters, spelling matters …
• This may seem complicated at first, but you will get the hang of it with a bit of practice!

An example: the seq() function

• Code completion is helpful here too: type se and hit Tab
• Select seq() by typing q or using the arrow keys
• Note that RStudio also gives you a bit of information about the function as you do this
• Once you have selected the right function, hit Tab again
• Note that RStudio already matches the opening and closing parentheses for you

seq(1,10)

 [1]  1  2  3  4  5  6  7  8  9 10

• Get help on this function by typing ?seq

Further assistance with writing code

• Type the following:

x <- "hello world"

• Note how RStudio again provides the matching second set of quotes for you
• You always need to have paired parentheses or quotes
• Otherwise, R will tell you that something is missing
  • try typing x <- "hello
  • the + shown in the console is the “continuation character”, telling you that something is missing
  • this is usually a " or a )
  • add the missing character and hit Enter (or press Esc to exit)

Replicability vs. reproducibility …

… and consequences for workflow

Replicability

A study is replicable if the same results are obtained when the study is repeated.

Reproducibility

An analysis is reproducible if the same results can be obtained from the data

• By documenting your analysis in an R script, you have already taken an important first step towards reproducibility.
• While it is perfectly ok to try out different things on the console (and this helps while you’re learning), get in the habit of pasting what works into a script to document your analysis.
• Better still: work directly from a Quarto document as we have been doing

Workflow

Different types of geoms and aesthetics

Aesthetics

• Note that you can add aesthetics either to the call to ggplot (left example), in which case they apply to all applicable geoms or to the geom itself (right example), in which case they only apply to that geom (more on this later)

penguins |> 
  ggplot(aes(x = flipper_length_mm, 
             y = body_mass_g, 
             colour = species)) +
  geom_point()

penguins |> 
  ggplot(aes(x = flipper_length_mm, 
             y = body_mass_g)) +
  geom_point(aes(colour = species))

Example 2: a categorical predictor

Is species predictive of body mass?

No data manipulation needed, body_mass_g (outcome) -> y-axis, species (predictor) -> x-axis. Try using geom_point() again …

penguins |> 
  ggplot(aes(x = species, 
             y = body_mass_g)) +
  geom_point() +
  labs(
    title = "Body mass by species",
    x = "Species",
    y = "Body mass (g)"
  )

The output is not quite what we would hope for …

Example 2a: a categorical predictor

Is species predictive of body mass?

For this type of question, geom_jitter() gives a better result.

penguins |> 
  ggplot(aes(x = species, 
             y = body_mass_g)) +
  geom_jitter() +
  labs(
    title = "Body mass by species",
    x = "Species",
    y = "Body mass (g)"
  )

Much better!

Example 2b: cat. predictor of means

Is species predictive of mean body mass?

Here, we can use geom_col(), but we need to do some summarising first.

penguins |> 
  group_by(species) |>  
  summarise(mean_body_mass_g = 
              mean(body_mass_g, na.rm=TRUE)) |>  
  ungroup() |>  
  ggplot(aes(x = species, 
             y = mean_body_mass_g)) +
  geom_col() +
  labs(
    title = "Body mass by species",
    x = "Species",
    y = "Mean body mass (g)"
  )

Don’t forget to adjust your aesthetics to reflect the new variable names!

Example 3a: geom_line without groups

How many penguins (of each species) were sighted across the different years?

penguins |>  
  count(year) |> 
  ggplot(aes(x = year, 
             y = n)) +
  geom_line() +
  labs(
    title = "Penguin count per year",
    x = "Year",
    y = "Number of penguin sightings"
  )

We will see how to fix the axis numbers later.

Example 3b: geom_line with groups

How many penguins of each species were sighted across the different years?

penguins |>  
  count(year,species) |>  
  ggplot(aes(x = year, 
             y = n,
             colour = species)) +
  geom_line() +
  labs(
    title = "Penguin count per year",
    subtitle = "and species",
    x = "Year",
    y = "Number of penguin sightings"
  )

Note the use of colour to introduce groups and also how we can easily add a subtitle.

Example 4: histograms

What is the distribution of bill depth?

We can use a histogram to answer this. Note that histograms only require an x aesthetic; the y-axis values are calculated automatically through the choice of geom. We also need to adapt the choice of binwith to suit the variable under consideration.

penguins |>  
  ggplot(aes(x = bill_depth_mm)) +
  geom_histogram(binwidth = 1) +
  labs(
    title = "Bill depth histogram",
    x = "Bill depth",
    y = "Count"
  )

Example 4a: histograms by group

What is the distribution of bill depth by species?

Pair a histogram with facets.

penguins |> 
  ggplot(aes(x = bill_depth_mm)) +
  geom_histogram(binwidth = 1) +
  labs(
    title = "Bill depth histogram by species",
    x = "Bill depth",
    y = "Count"
  ) +
  facet_wrap(~species)

Note the use of facet_wrap(~variable) to split the plot into facets according to the variable.

Example 5a: Boxplots with colour fill

What is the distribution of bill depth by species?

Note that colour is used for points, lines and outlines, while fill is used to fill shapes.

penguins |> 
  ggplot(aes(x = species, 
             y = bill_depth_mm, 
             fill = species)) +
  geom_boxplot() +
  labs(
    title = "Bill depth by species",
    x = "Species",
    y = "Bill depth"
  )

Example 5b: Boxplots with colour fill

What is the distribution of bill depth by species and island?

You can use fill to introduce an additional set of groups. Note also how you can change the label for a particular aesthetic using labs.

penguins |> 
  ggplot(aes(x = species, 
             y = bill_depth_mm, 
             fill = island)) +
  geom_boxplot() +
  labs(
    title = "Bill depth by species and island",
    x = "Species",
    y = "Bill depth",
    fill = "Island"
  )

The gapminder data

• Data from the Gapminder Foundation.
• Their mission is “to fight devastating ignorance with a fact-based worldview that everyone can understand” (from https://www.gapminder.org).
• They provide access to a wide variety of data about the world, including information on income, population and life expectancy across many countries.
• The {gapminder} package in R provides convenient access to an excerpt from the Gapminder data, focusing on life expectancy, GDP per capita and population by country.

• You will need to install the {gapminder} package by running install.packages("gapminder") in the console (don’t put this into your qmd document)

Do some simple exploration

• Inspect the data
• Ask some simple questions and create plots to answer them (keeping in mind that you may need to manipulate the data first)
• For example:
  • How has life expectancy in Germany changed over time?
  • What was the relationship between life expectancy and how wealthy a country is (gdpPercap) in 1952? In 2007?
  • How did the mean life expectancy per continent change over time?
  • Which continent had the highest variability in population across countries? Pick two years of your choice to examine. Convert population to millions before plotting.

Week 3 exercises

• Create a new RMarkdown document and use it to compile a summary of all the data manipulation functions (e.g. filter) and plot types (e.g. scatterplots) that we have looked at so far
• Don’t forget to load all the packages that you need first
• For each function or plot type write a brief description (1-2 sentences) of what the function / plot is used for. Provide one example question that you could answer using this function / plot on the basis of the gapminder and/or the palmerpenguins data and supply the necessary code to answer your question.
• Divide your document into appropriate subsections. Remember, the number of hashes specifies the level of subsection “##” for first level, “###” for the second level etc. Think about what goes together and how to best structure your document.
• Submit the knitted version via Ilias. (Hint: knit your document often as you’’re creating it to make any troubleshooting less difficult.)

Note: This document will also serve as a valuable reference for you going forward.